Download CMN 150 Exam 1 with correct answers and more Exams Project Management in PDF only on Docsity!
big |data |revolution |- |correct |answer |✔96% |mobile |phones |globally, |FB |2. |million |active |monthly |users, |twitter |330M |active |users, |thousands |of |satellites, |traffic |cams, |etc
most |commonly |use |digital |footprint |in |developing |countries |- |correct |answer
|✔mobile |phones |and |satellites
feature |engineering |- |correct |answer |✔convert |raw |quantitative |data |into |a |set |of |"features" |such |as |convey |phone |logs |into |"meaningful" |metrics
supervised |machine |learning |- |correct |answer |✔given |a |set |of |training |data |with |known |inputs |and |outputs, |learning |a |function |that |maps |an |input |to |outputs
deterministic |finite |automato |- |correct |answer |✔a |structured |way |of |recursively |generating |features; |simply |need |to |program |finite |state |machine |and |derive |these |metrics; |left |with |tens |and |thousands |of |metrics
traditional |programming |- |correct |answer |✔data |& |program |--> |computer |--> |output
machine |learning |- |correct |answer |✔data |& |output |--> |computer |--> |program
single |feature |most |correlated |with |wealth |- |correct |answer |✔weighted |average |of |all |first |degree |neighbors |"day |of |week |entropy" |of |outgoing |SMS |volume
band-aid |statistics |- |correct |answer |✔methods |that |produce |statistics |roughly |as |accurate |as |a |5-year |old |national |survey; |500x |cheaper, |20x |faster
real-time |monitoring |- |correct |answer |✔detecting |changes |in |welfare |which |can |enable |better |targeting |of |aid |and |assistance, |new |approaches |to |impact |evaluation, |and |disaster |response
digital |footprint |- |correct |answer |✔veracity; |automated |recording, |often |unintentional, |unavoidable |byproduct
data-fusion |- |correct |answer |✔variety; |messy |and |incomplete; |integrating |multiple |data |sources |to |produce |more |complete |data |source
no |sampling |- |correct |answer |✔volume; |big |but |biased
real-time |- |correct |answer |✔velocity; |dynamic, |often |quickly |available
machine |learning |- |correct |answer |✔automated |insights, |big |data |analytics, |makes |sense |of |digital |footprint |using |machine |algorithms
big |data |- |correct |answer |✔digital |footprint, |data-fusion, |no |sampling, |real- time, |machine |learning
OCEAN |- |correct |answer |✔openness, |conscientiousness, |extraversion, |agreeableness, |neuroticism
macro |level |example |of |real |time |- |correct |answer |✔U.S. |Bureau |of |labor |statistics; |more |than |hundreds |of |staff |in |over | 90 |cities;
17 |staff, | 300 |online |retailers |in |over | 70 |countries, |daily |fluctuations |of | 5 |million |prices, |data |is |more |fine-grained
micro |level |example |of |real |time |- |correct |answer |✔indiviual |person; |populations |driven |by |30% |emotions, |25% |thoughts, |20% |reactions, |10% |opinions, |10% |reflections, |5% |actions;
matching |personality |types |for |customer |service |and |satisfaction
content-based |filtering |- |correct |answer |✔learn |something |about |you |using |patterns |over |time, |personal |interacts |with |different |products, |spends |time |or |money |with |things, |need |time |and |history |of |a |person |to |make |better |recommendations
collaborative |filtering |- |correct |answer |✔uses |data |on |different |clients |to |predict |characteristics |of |other |clients
footprint |is |not |necessarily |representative |- |correct |answer |✔digital |divide |(which |keeps |growing); |i.e |men |are |more |represented |in |rwanda |although |population |is |50/
data |is |not |reality |example |- |correct |answer |✔predpol, |which |predicts |crime |in |real |time; |suggests |police |to |make |rounds |at |certain |times |at |certain |places; |leads |to |predictor |discrimination |and |killing |innocent |people
meaning |does |not |mean |something |is |meaningful |example |- |correct |answer
|✔associating |white |ethnic |origin |names |with |more |positive |words |and |african |american |origin |names |that |are |more |negative
discrimination |is |not |personalization |example |- |correct |answer |✔recommended |order |of |prices |of |the |same |thing |different |for |different |ethnicities; |princeton |review |charging |asians |higher |price |for |SAT |prep
past |is |not |the |same |as |future |example |- |correct |answer |✔data |science |may |not |be |able |to |predict |your |future |behavior |if |you |suddenly |fall |in |love, |change |your |job, |or |switch |countries |because |behaviors |will |change
clusters |of |obesity |- |correct |answer |✔three |degrees |of |separation
influencing |- |correct |answer |✔your |friend |does |something, |so |you |do |it |too
homophily |- |correct |answer |✔we |choose |friends |that |are |like |us
context |- |correct |answer |✔you |and |your |friend |are |influenced |by |a |third |thing
smoking |clusters |- |correct |answer |✔smokers |were |being |pushed |outside |social |clusters |as |smoking |became |less |popular
drinking |clusters |- |correct |answer |✔if |a |woman |stops |drinking, |more |friends |are |more |likely |to |stop |versus |men
transaction |theory |- |correct |answer |✔cost; |common |culture |("Weltanschauung") |facilitates |communication
dialectic |theory |- |correct |answer |✔social |pressure; |current |narrative/prejudices
evolution |theory |- |correct |answer |✔social |competition; |group/kin |selection |theory
multi-mode |networks |- |correct |answer |✔when |one |analyzes |a |network |with |different |kinds |of |nodes
new |kind |of |node |- |correct |answer |✔often |exclusive
attribute |- |correct |answer |✔very |overlapping
links |- |correct |answer |✔when |two |people |communicate |with |each |other
multiplex |networks |- |correct |answer |✔when |one |analyzes |a |network |with |different |kinds |of |links
social |evolutionary |pressure |- |correct |answer |✔selects |for |who |you |are |and |with |whom |you |connect
tie |strength |- |correct |answer |✔real |ties |are |in |grey |zone |between | 0 |and |1: |usually |implicit |or |explicit |dichotomization
points |- |correct |answer |✔vertices, |nodes, |sites, |actors
lines |- |correct |answer |✔edges, |arcs, |links, |bonds, |ties, |relations
isolate |- |correct |answer |✔node |that |is |not |connected
pendant |- |correct |answer |✔node |connected |to |one |node
dyad |- |correct |answer |✔two |connected |nodes
triad |- |correct |answer |✔three |connected |nodes
directed |network |- |correct |answer |✔communication, |directionality, |one |way, |asymmetric, |ie |Twitter, |one |person |does |one |thing, |doesn't |mean |the |other |person |is
undirected |network |- |correct |answer |✔friendship, |friends |both |ways, |symmetric, |ie |facebook, |both |people |are |doing |the |same |thing
degree |- |correct |answer |✔number |of |links |of |each |node, |both |outgoing |and |ingoing
one |link |- |correct |answer |✔has |two |degrees |and |connects |two |degrees
brokers |- |correct |answer |✔actors |who |exploit |structural |holes; |gain |access |to |info, |power |to |filter, |timing |for |competitive |advantage, |and |ability |to |refer |other |actors
structural |holes |- |correct |answer |✔separation |between |non-redundant |contacts; |places |where |people |are |unconnected |in |a |network
degree |centrality |- |correct |answer |✔most |connected; |has |lots |of |connections; |highest |degree |centrality
closeness |centrality |- |correct |answer |✔closest |to |all; |fewer |steps |to |all |others; |count |degrees |of |separation |and |measures |the |distance |between |nodes
betweeness |centrality |- |correct |answer |✔on |the |paths |connecting |all; |gatekeeper, |intermiediary, |or |broker; |sum |of |shortest |paths |through |node/all |shortest |paths; |counts |the |shortest |paths |among |nodes; |high |level |measures |which |nodes |build |bridges |or |are |in-between |others |in |community |structure
Girvan-Newman |Algorithm |- |correct |answer |✔calculates |the |betweeness |of |all |existing |ties, |remove |the |tie |with |the |highest |betweeness, |recalculate |betweeness |of |all |ties, |and |repeat |steps | 2 |and | 3 |until |no |ties |remain
eigenvector |centrality |- |correct |answer |✔proportional |to |sum |of |neighbor's |centrality; |measures |the |number |of |friends |of |node's |friends
global |measure |- |correct |answer |✔average |degree, |degree |distribution, |path |length, |etc; |gives |one |number |per |network
local |measures |- |correct |answer |✔clustering, |transitivity, |structural |equivalence, |etc; |provides |several |numbers |per |network
individual |measures |- |correct |answer |✔degree |centrality, |closeness, |betweeness, |eigenvector, |etc; |one |number |per |node
SNA |software |- |correct |answer |✔data |represented |by |nodes |and |links, |adjacency |matrix, |adjacency |list, |and |edge |list |that |contains |source |and |target
artificial |- |correct |answer |✔not |naturally |occurring; |created |by |humans
intelligence |- |correct |answer |✔rational; |deliberately |pursuing |goals, |human |like, |subjective
turing |test |- |correct |answer |✔a |method |of |determining |the |strength |of |artificial |intelligence, |in |which |a |human |tries |to |decide |if |the |intelligence |at |the |other |end |of |a |text |chat |is |human
AI |winter |- |correct |answer |✔period |of |reduced |funding |and |interest |in |artificial |intelligence |research; |disappointment |and |criticism, |followed |by |funding |cuts, |followed |by |renewed |interest |years |or |decades |later |in |the |1970s
AI |working |definition |- |correct |answer |✔consider |AI |challenges |as |those |that |can |be |done |by |humans, |but |not |yet |by |machines
aim |of |regression |- |correct |answer |✔predict |one |variable |on |basis |of |others |by |looking |for |a |shape |that |minimizes |the |errors/deviations |of |the |empirically |observed |points |from |the |idealized |shapes
overfitting |- |correct |answer |✔occurs |when |there |is |a |function |that |is |too |complex |and |is |too |closely |fit |to |a |limited |set |of |data |points
two |ways |to |get |around |overfitting |- |correct |answer |✔get |a |lot |of |data |or |apply |regulation |methods |to |force |the |fit |to |not |se |as |many |parameters
regularization |- |correct |answer |✔to |force |the |fit |not |to |use |parameters |more |than |it |needs
number |of |parameters |- |correct |answer |✔defines |the |complexity |of |the |model |used |by |ML
hyperparameters |- |correct |answer |✔limit |the |number |of |parameters |of |the |model, |and |therefore |the |complexity |of |the |model
testing |set |- |correct |answer |✔used |to |identify |how |accurate |the |decision |rule |is |for |unseen |causes
training |set |- |correct |answer |✔used |to |identify |a |pattern |in |the |data
validation |set |- |correct |answer |✔used |to |decide |which |kind |of |decision |rule |generalizes |best
election |forensics |- |correct |answer |✔development |of |tools |for |exploring |electoral |data |looking |for |potential |indicators |of |electoral |manipulation
ballot |box |stuffing |- |correct |answer |✔occurs |when |illegitimate |votes |are |cast |by |non-existent |voters |or |when |more |than |one |vote |is |cast |per |voter
symptoms |of |ballot |box |stuffing |- |correct |answer |✔inflated |turnout |rates; |inflated |voter |shares |favoring |one |party |or |candidate;
ballot |box |stuffing |model |- |correct |answer |✔there |will |be |places |where |voter |turnout |is |unusually |large |and |will |see |a |multimodal |distribution
vote |stealing |- |correct |answer |✔occurs |when |votes |are |stolen |from |one |party |and |either |destroyed |or |transferred |to |another |party
symptoms |of |vote |stealing |- |correct |answer |✔deflated |vote |shares |for |one |party; |inflated |shares |for |another |party; |if |stolen |votes |are |destroyed, |deflated |turnout |rates
vote |stealing |model |- |correct |answer |✔same |amount |of |voter |turnout |but |lower |opposition |vote |share |thane |expected, |and |for |the |incumbent |party, |there |is |higher |incumbent |vote |share
supervised |learning |approach |- |correct |answer |✔data |is |classified |into |clean |or |tainted |based |on |resemblance |to |clean |and |tainted |cases |in |a |training |data |set;
kant's |free |will |- |correct |answer |✔if |we |have |free |will |then |why |can |we |still |make |predictions |about |people
three |legs |of |social |science |- |correct |answer |✔empirical, |theoretical, |analytical
empirical |- |correct |answer |✔observe, |data |collection, |reality, |big |data, |darwin
theoretical |- |correct |answer |✔world |of |thoughts, |theories |about |societies, |computer |simulations, |einstein
analytical |- |correct |answer |✔how |something |happened
induction |- |correct |answer |✔data |--> |theory
deduction |- |correct |answer |✔hypothesis |--> |data
red |wine |theorizing |- |correct |answer |✔create |a |theory |--> |verbally |deduce |a |hypothesis, |then |collect |data |to |reject |or |accept
replication |crisis |- |correct |answer |✔half |of |what |we |know |about |society |is |wrong |but |we |don't |know |which |half
incompleteness |theorem |- |correct |answer |✔no |one |will |ever |be |able |to |show |that |this |is |the |best |way |of |doing |things
entropy |- |correct |answer |✔variance |of |distribution
example |of |voices |- |correct |answer |✔radio |stations, |protest |leaders
example |of |amplifiers |- |correct |answer |✔celebrities, |mediators |in |news, |own |voices
weak |component |- |correct |answer |✔non-directed |path |between |every |pair
strong |component |- |correct |answer |✔every |node |is |reachable |from |every |other |node
